2388
PILGRAM AND PHILLIPS
VOL.30
Reaction of Aromatic Thiocyanates with Trialkyl Phosphites KURTPILGRAM AND DONALD D . PHILLIPS' The Agricultural Research Division, Shell Development Company, A Division of Shell Oil Company, Modesto, California Received November 27, 1964 In contrast to aliphatic thiocyanates, which react with trialkyl phosphites exclusively by an Arbuzov mechanism to form phosphorothioates, aromatic thiocyanates, ArSCN, react with trialkyl phosphites, ( RO)3P, to give a mixture of phosphorothioate and sulfide, RSAr. The formation of sulfide is favored by substituents (NO?, etc.) on the aryl ring that can stabilize an anionic species, an observation that suggests a mechanism involving attack of phosphite on the carbon of the thiocyano group followed by alkylation of the resulting thiophenoxide ion by the phosphonium cation. The reactions of trialkyl phosphites with related sulfur compounds, RSX (disulfides, sulfenyl chlorides, sulfenate esters, etc.), are rationalized on the theory that heterolysis of the RS-X bond will occur in the direction RS+X- when X - is the anion of a fairly strong acid.
The classical Rilichaelis-Arbuzov reaction between trivalent phosphorus compounds and alkyl halides has been considerably expanded in scope since its discovery2 and now embraces as coreactants for the trivalent phosphorus compound a large number of compounds containing displaceable groups other than halogen. Among these coreactants are a substantial number of sulfur compounds such as sulfonyl halides, sulfenyl halide^,^ thiosulfonates,6 symmetrical' and unsymmetricals disulfides, and thiocyanate^.^ Work on the reaction of thiocyanates with trialkyl phosphites has been quite sketchy; one groupga investigated the reactions of two trialkyl phosphites with ethyl and benzyl thiocyanate and came to the conclusion that the reactions followed the typical Michaelis-Arbuzov course (eq. 1). SheppardQcpostu)p4
0
t
(R0)zPSR'
+ RCN
"Q~oz II
lated the same mechanism independently and, although he extended the reaction to include two aryl thiocyanates, no departure from the expected Arbuzov products was observed. When the reaction was extended by one of uslo to nitroaryl thiocyanates, however, it was found to take a different course in that the main product containing the nitroaryl moiety was not the phosphorothioate (I) but was instead the alkyl nitroaryl sulfide (11) (eq. 2). The fate of the phosphorus compound (111) was not established with certaint'y but ( 1 ) To whom inquiries regarding this article should be addreseed at Mobil Chemical Research, Edison Township Laboratory, Metuchen, N. J. (2) G . M . Kosolapoff, "Organophosphorus Compounds." John Wiley and Sons, Inc., New York, N. Y . , 1950, Chapter 7 : P. C. Crofts. Quart. Rev. (London), 12, 341 (1958). (3) E . E . Gilbert and C. J. McGough, U. S. Patents 2,690,450 and 2,690,451 (1954). (4) F. W . Hoffmann, T . R. Moore, and B. Kagan, J . A m . Chem. Soc., 78, 6413 (1956). (5) D. C. Morrison, ibid., 7 7 , 181 (1955). (6) (a) T. C. Myers, S. Preis, and E. V. Jensen, ibid.. 76, 4172 (1954); (b) J. Michalski, T. Modro, and J. Wieczorkowski, J . Chem. Soc., 1665 (1960). (7) (a) H. I. Jacobson, R . G . Harvey, and E. V. Jensen, J . A m . Chem. Soc.. 7 7 , 6064 (1955); (b) .4. C.Poshkus and J . E. Herweh, ibid., 79, 4245 (1957); (c) J. Michalski and J . Wiecaorkowski, Bull. acad. polon. sci., Classe III, 6 , 917 (1957); Chem. Abstr., 63,6157 (1958); (d) C.Walling and R . Rabinowitz. J . A m . Chem. Soc.. 79, 5326 (1957); 81, 1243 (1959); (e) A . J . Parker and N . Kharasch, Chem. Rev., 69, 621 (1959). (8) R. G. Harvey, H . I. Jacobson. and E. V . Jensen, J . A m . Chem. Soc., 86, 1618 (1963). (9) (a) J. Michalski and J . Wieczorkowski, Bull. acad. polon. sci., Classe I I I , 4 , 279 (1956); Chem. Abstr.. 61, 4266 (1957); (b) K. Pilgram, P h . D . Thesis, Universitat Wursburg, 1959; (c) W. A . Sheppard, J . Ore. Chem.. 26, 1460 (1961). (10) K. Pilgram and F. Korte. Tetrahedron. 20, 177 (1964).
t
(2)
m
the detection of isonitrile suggested that a t least part of the phosphorocyanidate (111) reacted with excess trialkyl phosphite according to eq. 3. 0
(1)
0 4- [(RO),PCN]
t (ROLPCN I11
0 0
t + (RO)aP +RNC + (RO)2h-P(OR)2
(3)
At first, this apparently anomalous result with nitroaryl thiocyanates was ascribed to the unique electronwithdrawing properties of the nitro group,lO but it soon became clear from a qualitative study of several halogenated aryl thiocyanates that all of them gave sulfide in varying amounts when allowed to react with trimethyl phosphite. Consequently, in order to gain an insight into the mechanism of this "abnormal" reaction (eq. 2), we have studied the behavior of eleven halogenated aryl thiocyanates in their reaction with trimethyl phosphite. Relative reaction rates have been measured a t 60' and the product ratio (it?.,sulfidephosphorothioate) has been determined, both by gasliquid chromatographic (g.1.c.) techniques. Experimental Preparation of Aromatic Thiocyanates.-All of the thiocyanates used in this study (Table I ) were prepared from the corresponding aniline by diazothation followed by treatment with cuprous thiocyanate and potassium thiocyanate." Relative Reaction Rates.-The apparatus used to determine the relative reaction rates consisted of a four-necked flask equipped with a stirrer, thermometer, condenser, and sampling syringe. The experiments were run under nitrogen. The flask containing trimethyl phosphite (62 ml., 0.525 mole) was immersed to a depth above the reaction level in a water bath a t 62". The thiocyanate (0.025 mole) was added to the phosphite ester and, since the addition of the thiocyanate lowered the temperature of the phosphite by 2O, the bath temperature was also immedi(11) L. Gattermann and W . Hausknecht. Ber., PS, 738 (1890).
AROMATIC THIOCYANATES AND TRIALKYL PHOSPHITES
JULY1965
2389
TABLE I AROMATICTHIOCYANATES M.p. or b.p. (mm.), OC.
Yield, Structure
%
Formula (mol. wt.)
C7HaClzNS(204.0)
-Nitrogen, Calcd.
36 109 (0.1) C7HaClzN S (204.0) 68.8 60 l-SCN-3,4-C12CeHa 79 72 1-SCN-2,4-C1zCeHa C7HaClzNS (204.0) C7H3CLNS (204.0) 54 81 l-SCN-2,5-C1L!J& C7HaClNS (169.5) 40 126 (13) l-SCN-2-ClCtjH4' 65 134 (15) l-SCN-3-ClCeH4" C7HrClNS (169.5) C7HdFNS (153 1) 32.6 74 (0.01) l-SCN-2-FCJIa C7IIaFNS (153.1) 26.5 61 (0.05) l-SCN-3-FCeH4 CaHaFaNS (203.1) 57.3 64 (0.09) l-SCN-2-CFaCeHa CsHaF3NS (203.1) 48.8 58 (0.01) l-SCN-3-CFaCeH4 l-SCN-3-CFa-4ClCeHa CsHaClFaNS (237.6) 21 89 (0.1) a F. Challenger, C. Higginbottom, and A. Huntington, J . Chem. Soc., 29 (1930).
l-SCN-3,5-C4CsHg
u
h j z , min.
l-SCN-2-CFaCeH4 1-SCN-2-ClCsHr l-SCN-3-FCeH4 l-SCN-3-CICeH4 l-SCN-2,4-C12CeHg l-SCN-3-CFaCsHd l-SCN-2,5-ClzCeHs l-SCN-3-CFa-CClCsHs l-SCN-2-FCeH4 l-SCN-3,5-C12CeHa l-SCN-3,PClzCsHg
... 0.20 0.34 0.37 0.43 0.42 0.57 0.74 0.24 0.75 0.60
PRODUCT RATIOFOR
THE
6.8
15.7
15.3
C1,34.8 c1, 34.8 c1, 34.8 C1, 20.9 C1, 20.9 9.2 9.2 6.9 6.9 5.9
C1, 35.1 c1, 34.7 C1, 3 4 , 6 C1, 20.9 C1, 21.2 8.9 8.9 7.0 6.8 5.6
15.7 15.7 15.7 18.9 18.9 20.9 20.9 15.8 15.8 13.5
15.4 15.8 15.3 18.8 18.8 21.2 21.2 15.6 15.6 12.8
TABLE I11 AROMATICTHIOCYANATE-TRIMETHYL PHOSPHITE REACTION
Temp., 60°-
-
Composition after heating
I
0 RSCN R
Sulfur, %---Calcd. Found
_ .
usually identified by preparing the sulfide and phosphorothioate independently on a macroscale and analyzing them by g.1.c. under the same conditions. In all cases the order of emergence was sulfide, thiocyanate, and lastly phosphorothioate. I n this way the stability of the phosphorothioate to the analytical method was verified. After all of the thiocyanate had disappeared and the ratio of sulfide to phosphorothioate had remained constant for a t least 2 hr., the reaction mixture was heated to reflux (115-120°) for different periods of time and then analyzed for sulfide and phosphorothioate by the same g.1.c. procedure. I n some cases, the reaction mixture was then heated from 150-180' and, after all the trimethyl phosphite had been isomerized to dimethyl methylphosphonate, the sulfide-phosphorothioate ratio was again determined by g.1.c. This was done because in our earlier experiments with some of these thiocyanates, a different ratio of products was obtained a t higher temperatures than a t 60'. The results are listed in Table 111.
(Hammett) or u* (Taft)
133 92 90 80 66 63 53 52 50 38 36
Found
6.9
16.7
TABLE I1 HALF-LIVES A N D HAMMETT u VALUESFOR THE REACTION OF AROMATIC THIOCYANATES WITH TRIMETHYL PHOSPHITE Thiocyanate
%------
0
t
t
RSCH:, %
RSP(OCHs)z, %
RSCHI. %
RSP(OCH:)a, %
Time, min.
Temp., OC.
2-FCsHc 3-FCsHr 3-CFpCeHr
8.3 10.2 14.9
91.7 89.8 85.1
2-CFaCeH4-
25.3
74.7
2-CICeH43-CFa-4ClCeHp 3-ClCeH4-
28 29.3 32.2
72 70.7 64.8
37.3
62.7
89 87.7 84.8 0 72.1 71.5 72.4 62.8 67.5 52.3 48.7 5.5 35 20 0 0
60 30 10 120 10 120 10 45 90 60
3,4-C12CeHg-
11 12.3 15.2 100 27.9 28.5 27.6 37.2 32.5 47.7 51.3 94.5 65 80 100 100
115 115 115 180 115 115 115 115 115 150 115 180 140 180 180 180
2,4-ClzCeHa3,5-ClzCsHa-
41.8 46.7
58.2 53.3
2,5-C1zCeHa-
65
35
ately lowered to 60 f 0.5' by adding water. (A few experiments were carried out a t 50 f 0.5O.) At fixed time intervals, 0.5-ml. samples of the reaction mixture were withdrawn and analyzed by injecting 40 ml. into a Model 500 linear programmed temperature gas chromatograph ( F and M Scientific Corp., Avondale, Pa.). The 2-ft. column was packed with 30-60-mesh Chromosorb P, coated with 2074 silcone rubber; the injection port temperature was 200". The relative reaction rates (shown in Figures 1-4 and expressed as half-lives in Table 11) were determined by following the disappearance of thiocyanate, and the product ratios shown in Table I11 were determined from the relative peak areas assignable to sulfide and phosphorothioate, respectively. These peaks were
io 30 60 45 60 60
Because the reaction proceeds to a small extent immediately on mixing a t 60' and, because the injection temperature of 200' results in some reaction, none of the curves passes through 100% for thiocyanate or 0% for sulfide or phosphorothioate a t zero time. Since all of the thiocyanate-trimethyl phosphite reactions were analyzed by the same procedure, however, it was felt that any error introduced by the analytical method would be small. The extremely good reproducibility of the results (1-2%) tended to confirm this. Preparation of Methyl n,a,a-Trifluoro-o-tolyl Sulfide .-This and the following experiment illustrate a typical macroscale preparation of the two products from the reaction of trimethyl
PILGRAM AND PHILLIPS
2390
VOL.30
reactions : one presumably involves attack on sulfur to form phosphorothioate (eq. 1) and the other presumably (see below) involves attack on carbon to form aryl sulfide (eq. 4). Furthermore, the reactions were
100
90
(R0)sP
+ ArSCN + [(R0)3PtCN ArS-] + 0
'.
\
\ 10
2
I
3 rlmr imri
4
5
Figure 1.-Reaction of 2,4-dichlorophenyl thiocyanate with trimethyl phosphite; rate of thiocyanate consumption a t 50".
ArSR
t + [(RO)zPCN]
(4)
run pseudo first order, i e . , in the presence of a large excess of trimethyl phosphite. Changes in the composition of the reaction mixture by formation of products and isomerization of trimethyl phosphite caused a tailing off (Figure 1). In spite of these complications it seems clear from the data in Table I1 that both reactions are favored by electron-withdrawing groups in the aryl ring. For example, among the meta-substituted thiocyanates in Table I1 the reactivity parallels the Hammett (T values. Similarly, the 2,5-dichloro isomer ((T a*I3 =
+
O S C N
m, t,,~, min. u
Figure 2.-Reaction of 2,4-dichlorophenyl thiocyanate with trimethyl phosphite a t 50'. phosphite with an aromatic thiocyanate, i.e., the methyl aryl sulfide and the phosphorothioate. To 33 g. (0.162 mole) of a,a,a-trifluoro-o-tolyl thiocyanate in 50 ml. of methanol was added, with stirring, a solution of 30 g. of sodium hydroxide in 150 ml. of methanol. The cold solution ( 0 ' ) of the resulting mercaptide was treated with 25 g. (0.2 mole) of dimethyl sulfate and stirred for an additional 2 hr. a t room temperature. The reaction mixture was diluted with water, extracted with ether, and the dried ethereal extract evaporated a t reduced pressure. The pale yellow oil was distilled to give 26 g. (83%) of the sulfide: b.p. 102" (22 mm.), m.p. 12", n z 5 1.5088; ~ lit.1zm.p. 3.7-5", b.p. 96" (25 mm.), n25~1.509. A n d . Calcd. for CaH7FaS: S, 16.65. Found: S, 17.0. The sulfone crystallized from methanol as colorless needles, m.p. 75'. Anal. Calcd. for CSH7F30zS: S, 14.28. Found: S, 14.1. Preparation of 0,O-Dimethyl S-a,a,a-Trifluoro-o-tolyl Phosphorothioate.-A solution of 30.5 g. (0.15 mole) of a,a,a-trifluoro-o-tolyl thiocyanate in 20 g. (0.16 mole) of trimethyl phosphite was heated slowly to 95' a t which point a vigorously exothermic reaction ensued and the temperature briefly rose to 200'. The reaction product was distilled in vacuo to give 18 g. of a forerun, b.p. -7"
HCN 9 ,3ZC
+
HSEt
HOCH:
12.0d
17'
OT.G. McCoubert, Trans. Faraday SOC.,51, 743 (1955).
* F. G. Bordwell and H.
M. Anderson, J . Am. Chem. SOC.,7 5 , 6019 (1953). e H. T. S. Britton and R. A . Robinson, Trans. Faraday SOC.,28, 531 (1932). J. Maurin and R. A. Paris, Compt. rend., 232, 2428 (1951). e N. Bjerrum, A. Ungemach, and L. Zechmeister, Kgl. Danske Videnskab. Selskab. Overs. Selskab. Virksomhed, 5 , 1 (1924).
acid, HX. For example, sulfenyl halides, ArSCl, react with trialkyl phosphites exclusively by heterolysis in the direction shown in Table IV; mixed disulfides, RSSAr, also react exclusively in this direction*; aryl thiocyanates, ArSCN, react partly in the direction shown and partly by heterolysis in the opposite direction (this work); mixed disulfides, ArSSR, react exclusively in the opposite directions; and alkyl sulfenate esters, ArS@CH3,fail to react a t all. A final observation from Figures 2-4 and Table I11 pertains to the dramatic change in composition that occurs when the mixture of sulfide and phosphorothioate is heated to 140-180'. In all cases the sulfide concentration increases a t the expense of the phosphorothioate; the extreme may be seen in Figures 3 and 4 where 2 hr. of heating a t 180' has converted all of the phosphorothioate to sulfide. We regard this reaction as a thermal alkylation (eq. 8) which is not 0
L + A
( RO), SAr
(A
RSAr t
-01')
(8)
directly related to the thiocyanate-trialkyl phosphite reaction because it occurs at a measureable rate only a t elevated temperatures. Structural effects on the ease of this reaction have been reported in a separate communication. l9 Acknowledgment.-The authors are grateful to Dr. C. H. Tieman, of this laboratory, for helpful discussions. (19) K . Pilgram and F. Korte. Tefrahedron,31, 203 (1965)
